Clinical characteristics of Candida tropicalis fungaemia with reduced triazole susceptibility in Taiwan: a multicentre study.

Candida tropicalis is the second most common Candida species causing fungaemia in Taiwan, and decreased susceptibility to fluconazole has been reported. This study analysed the clinical characteristics of adult patients with C. tropicalis fungaemia and the antifungal susceptibilities of isolates at five tertiary hospitals in Taiwan (1 July 2011 to 30 June 2014). A standardised case record form was used retrospectively to collect demographic, clinical and microbiological characteristics, antifungal treatment and outcomes. MICs of non-duplicate isolates were determined using SensititreTM YeastOneTM and were interpreted using cut-off values recommended by the CLSI. A total 248 patients were diagnosed over the study period; 30-day crude mortality was 52.0%. Multivariate analysis showed that high Charlson comorbidity index ≥4 (OR = 2.09, 95% CI 1.22-3.59; P = 0.008), neutropenia (OR = 4.61, 95% CI 1.42-15.00; P = 0.011) and treatment with an azole-based regimen (OR = 0.39, 95% CI 0.17-0.90; P = 0.028) were significantly associated with 30-day mortality. A total of 33.9% of isolates were non-susceptible to fluconazole (MIC50, 2 mg/L; MIC90, 16 mg/L; MIC range, 0.25 to >256 mg/L), whilst 56.9% to voriconazole (MIC50, 0.25 mg/L; MIC90, 1 mg/L; MIC range, 0.015 to >8 mg/L) according to CLSI clinical breakpoints. There was no significant correlation between overall mortality and MICs of fluconazole or voriconazole. This study showed high mortality in patients with C. tropicalis fungaemia, and azole-based antifungal treatment could improve outcomes regardless of fluconazole MICs of infecting isolates compared with patients without any treatment within 48 h.

Occult Klebsiella pneumoniae bacteremia at emergency department: A single center experience.

BACKGROUND/PURPOSE: Patients with undetected bacteremia when discharged from a hospital are considered to have occult bacteremia. Klebsiella pneumoniae bacteremia (KPB) is endemic to Taiwan. Our purpose was to study the impact of occult KPB. METHODS: We retrospectively reviewed the records of patients who were discharged from our emergency department (ED) and subsequently diagnosed with KPB (occult bacteremia), from January 2008 to March 2014. All patients are followed for at least 3 months after the index ED visit. The study group was compared to KPB patients who were directly hospitalized (DH) from ED in 2008. Thirty-day mortality was the primary endpoint. RESULTS: A total of 913 patients were admitted to our ED with KPB, and 88 of these patients (9.6%) had occult KPB. Among them, 43 had second ED visit and 41 were admitted. The overall 30-day mortality was 2.3%. Relative to patients with occult KPB, DH patients had more respiratory tract infections (p < 0.001) but fewer other intra-abdominal infections (p = 0.015). Liver abscess was the major diagnosis for the second ED visit (37.2%). DH patients had significantly greater 30-day mortality than that of overall patients with KPB (19.2% vs.2.3%, p < 0.001). CONCLUSION: Most patients with occult KPB had favorable outcomes, but about half of them required a second ED visit. Clinicians should aggressively follow patients with occult KPB and should seek to identify the focus of infection in this endemic area.

Increased rifampicin resistance in blood isolates of meticillin-resistant Staphylococcus aureus (MRSA) amongst patients exposed to rifampicin-containing antituberculous treatment.

The aim of this study was to determine the rifampicin (RIF) resistance rate of meticillin-resistant Staphylococcus aureus (MRSA) amongst patients with MRSA bacteraemia who have or have not been exposed to RIF-containing antituberculous (anti-TB) treatment. From 2000 to 2008, patients with MRSA bacteraemia and previous exposure to RIF-containing anti-TB therapy were selected. Patients matched for sex, age and time of culture of MRSA bacteraemia but without exposure to anti-TB therapy were selected as a control group. A total of 139 patients, comprising 49 with RIF exposure and 90 without RIF exposure, were analysed. The RIF resistance rate was higher in patients with previous RIF exposure (61.2% vs. 20.0%; P<0.001). The minimum inhibitory concentration of RIF that inhibited 50% of MRSA isolates (MIC(50)) for the study group was also higher (128 mg/L vs. 0.015 mg/L; P<0.001). The mortality rate was higher in the study group (59.2% vs. 41.1%; P=0.041). MRSA isolates recovered from patients with current usage of a RIF-containing anti-TB regimen were more likely to be resistant to RIF (87.5% vs. 36%; P=0.001), with higher MIC(50) values (256 mg/L vs. 1mg/L; P=0.002), and resulted in a higher mortality rate than isolates from patients with remote usage of an anti-TB regimen (79.2% vs. 40%; P=0.005). Multivariate analysis showed that current anti-TB drug usage was the only risk factor for RIF resistance [odds ratio (OR)=7.457, 95% confidence interval (CI) 1.581-35.167] and mortality (OR=7.201, 95% CI 1.583-32.766). Given the high rate of RIF resistance in patients with prior anti-TB treatment, RIF susceptibility testing should be performed before considering combination treatment of RIF in MRSA infection.

Correlation between antibiotic consumption and resistance of Gram-negative bacteria causing healthcare-associated infections at a university hospital in Taiwan from 2000 to 2009.

OBJECTIVES: This study investigated the correlation between antibiotic consumption and antimicrobial resistance in Gram-negative bacteria causing healthcare-associated infections at a university hospital in Taiwan from 2000 to 2009. METHODS: Disc susceptibility data of Escherichia coli, Klebsiella pneumoniae, Enterobacter cloacae, Serratia marcescens, Proteus spp., Pseudomonas aeruginosa, Acinetobacter spp., Stenotrophomonas maltophilia and other non-fermentative Gram-negative bacilli causing healthcare-associated infections were evaluated. Data on annual patient-days and annual consumption (defined daily doses per 1000 patient-days) of extended-spectrum cephalosporins, ß-lactam/ß-lactamase inhibitor combinations, carbapenems, aminoglycosides and fluoroquinolones were analysed. RESULTS: The trend of total consumption of extended-spectrum cephalosporins, ß-lactam/ß-lactamase inhibitor combinations, carbapenems, aminoglycosides and fluoroquinolones significantly increased between 2000 and 2003 and remained stable between 2004 and 2009. The decreasing use of gentamicin and amikacin in recent years was associated with increasing susceptibility of E. coli, E. cloacae, S. marcescens and P. aeruginosa to gentamicin, as well as increasing susceptibility of P. aeruginosa to amikacin. The use of piperacillin/tazobactam was positively correlated with the prevalence of piperacillin/tazobactam-resistant E. coli and S. maltophilia. In contrast, the use of cefotaxime and piperacillin/tazobactam was negatively correlated with the prevalence of cefotaxime-resistant E. coli and piperacillin/tazobactam-resistant S. maltophilia, respectively. The consumption of fluoroquinolones was positively correlated with the rates of ciprofloxacin-resistant E. coli, piperacillin/tazobactam-resistant P. aeruginosa and ceftazidime-resistant S. maltophilia. CONCLUSIONS: The relationship between antibiotic prescription and the rates of resistance for Gram-negative bacteria is complicated; every type of antimicrobial agent or even individual agent can have distinct associations with different pathogens.

Risk of recurrent nontyphoid Salmonella bacteremia in HIV-infected patients in the era of highly active antiretroviral therapy and an increasing trend of fluoroquinolone resistance.

BACKGROUND: Risk of recurrent nontyphoid Salmonella (NTS) bacteremia and trends of antimicrobial resistance of NTS remain unknown in human immunodeficiency virus (HIV)-infected patients receiving highly active antiretroviral therapy (HAART). METHODS: Ninety-three patients who received a diagnosis of NTS bacteremia from June 1994 through June 2006 were prospectively followed up. Incidence of recurrent NTS bacteremia was compared between the pre-HAART era (June 1994-March 1997) and the HAART era (April 1997-June 2006). Prevalence of antimicrobial resistance was compared among the NTS isolates obtained in the pre-HAART era, the early HAART era (April 1997-June 2002), and the late HAART era (July 2002-June 2006). RESULTS: Compared with patients enrolled in the pre-HAART era, patients who received HAART had an incidence of recurrent NTS bacteremia that was significantly reduced by 96%; the incidence of recurrent NTS bacteremia was 2.56 cases per 100 person-years in the HAART era, compared with 70.56 cases per 100 person-years in the pre-HAART era (rate ratio, 0.036; 95% confidence interval, 0.012-0.114; P<.001). In the HAART era, the incidence of recurrent NTS bacteremia did not increase among patients receiving fluoroquinolone prophylaxis for 30 days (3.95 cases per 100 person-years), with a rate ratio of 0.43 (95% confidence interval, 0.07-2.58). Although resistance to ampicillin, cotrimoxazole, and chloramphenicol decreased, the proportion of NTS isolates resistant to fluoroquinolones increased from 0% in the pre-HAART era to 6.2% in the early HAART era and 34.2% in the late HAART era (P=.002). CONCLUSIONS: The risk of recurrent NTS bacteremia decreased significantly in the HAART era, although NTS isolates obtained from HIV-infected patients were increasingly resistant to fluoroquinolones.